Emulator-Archive/ppsspp
1
0
Fork 0
mirror of https://github.com/hrydgard/ppsspp.git synced 2025-02-25 16:41:04 +00:00
Code Issues Actions 2 Packages Projects Releases Wiki Activity

Merge pull request #722 from unknownbrackets/jit-vfpu

Browse Source 
Fix an overlap issue with vdot
This commit is contained in:
Henrik Rydgård 2013-02-17 01:59:26 -08:00
commit 04c27cbe73
3 changed files with 17 additions and 14 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

1
Core/MIPS/MIPSIntVFPU.cpp
View File

@ -936,6 +936,7 @@ namespace MIPSInt
}
d = sum;
ApplyPrefixD(&d,V_Single);
// TODO: Shouldn't this respect the mask?
V(vd) = d;
PC += 4;
EatPrefixes();

2
Core/MIPS/x86/CompFPU.cpp
View File

@ -87,7 +87,7 @@ void Jit::Comp_FPU3op(u32 op)
}
}
static u32 GC_ALIGNED16(ssLoadStoreTemp[1]);
static u32 GC_ALIGNED16(ssLoadStoreTemp);
void Jit::Comp_FPULS(u32 op)
{

28
Core/MIPS/x86/CompVFPU.cpp
View File

@ -146,24 +146,29 @@ void Jit::ApplyPrefixD(const u8 *vregs, u32 prefix, VectorSize sz, bool onlyWrit
// Vector regs can overlap in all sorts of swizzled ways.
// This does allow a single overlap in sregs[i].
bool DestRegOverlaps(int dreg, int di, int sn, u8 sregs[], int tn, u8 tregs[])
bool IsOverlapSafeAllowS(int dreg, int di, int sn, u8 sregs[], int tn, u8 tregs[])
{
for (int i = 0; i < sn; ++i)
{
if (sregs[i] == dreg && i != di)
return true;
return false;
}
for (int i = 0; i < tn; ++i)
{
if (tregs[i] == dreg)
return true;
return false;
}
// Hurray, no overlap, we can write directly.
return false;
return true;
}
static u32 GC_ALIGNED16(ssLoadStoreTemp[1]);
bool IsOverlapSafe(int dreg, int di, int sn, u8 sregs[], int tn, u8 tregs[])
{
return IsOverlapSafeAllowS(dreg, di, sn, sregs, tn, tregs) && sregs[di] != dreg;
}
static u32 GC_ALIGNED16(ssLoadStoreTemp);
void Jit::Comp_SV(u32 op) {
CONDITIONAL_DISABLE;
@ -335,18 +340,15 @@ void Jit::Comp_VDot(u32 op) {
int n = GetNumVectorElements(sz);
X64Reg tempxreg = XMM0;
if (!DestRegOverlaps(dregs[0], 0, n, sregs, n, tregs))
if (IsOverlapSafe(dregs[0], 0, n, sregs, n, tregs))
{
fpr.MapRegsV(dregs, V_Single, MAP_NOINIT);
tempxreg = fpr.VX(dregs[0]);
}
// Need to start with +0.0f so it doesn't result in -0.0f.
MOVSS(tempxreg, M((void *) &zero));
MOVSS(XMM1, fpr.V(sregs[0]));
MULSS(XMM1, fpr.V(tregs[0]));
ADDSS(tempxreg, R(XMM1));
for (int i = 1; i < n; i++)
for (int i = 0; i < n; i++)
{
// sum += s[i]*t[i];
MOVSS(XMM1, fpr.V(sregs[i]));
@ -357,7 +359,7 @@ void Jit::Comp_VDot(u32 op) {
if (!fpr.V(dregs[0]).IsSimpleReg(tempxreg))
{
fpr.MapRegsV(dregs, V_Single, MAP_NOINIT);
MOVSS(fpr.V(dregs[0]), XMM0);
MOVSS(fpr.V(dregs[0]), tempxreg);
}
// TODO: applyprefixD here somehow (write mask etc..)
@ -427,7 +429,7 @@ void Jit::Comp_VecDo3(u32 op) {
X64Reg tempxregs[4];
for (int i = 0; i < n; ++i)
{
if (DestRegOverlaps(dregs[i], i, n, sregs, n, tregs))
if (!IsOverlapSafeAllowS(dregs[i], i, n, sregs, n, tregs))
{
// On 32-bit we only have 6 xregs for mips regs, use XMM0/XMM1 if possible.
if (i < 2)